Malaria incidence in areas targeted for a cluster randomised controlled trial to prevent malaria in Ethiopia: Results from a pilot study
Taye Gari1, Eskindir Loha1, Wakgari Deressa 2, Alemayehu Hailu 2, Oljira Kenea 2, Meshesha Balkew2, Teshome Gebremichael 2, Bjarne Robberstad3, Hans J Overgaard 4,5, Bernt Lindtjørn 3
1 Hawassa University, Hawassa, Ethiopia. 2 Addis Ababa University, Addis Ababa, Ethiopia. 3University of Bergen, Bergen, Norway. 4Norwegian University of Life Sciences, Ås, Norway. 5 Institut de Recherche pour le Développement (IRD), MIVEGEC, Montpellier, France.
MalTrials project funded by RCN GLOBVAC, Norway (project no. 220554)
Objective A cluster randomised controlled trial will be carried out in central Ethiopia to evaluate the effect of combining indoor residual spraying (IRS) and long-lasting insecticidal treated nets (LLIN) on reducing malaria incidence. A pilot study was conducted to assess the variation in malaria episodes within and among villages. Information from this pilot study will serve as the baseline and for calculation of a sufficient sample size to detect significant differences in malaria episode between study arms of the trial.
Method A cohort of 5309 residents from 994 households in 29 villages were followed from August to December, 2013 in Adami Tulu district in central Ethiopia. The villages were selected based on average distance from Ziway lake shore where the majority of malaria mosquito breeding sites are located. A household census was done, followed by weekly home visits to collect malaria related data. Blood samples were taken from febrile patients. Microscopic slide examination was performed for malaria parasite identification. Incidence rate of malaria episodes was calculated.
Results Only 271 (27.3%) of the households owned at least one insecticidal treated net at the start of the pilot study. The average proportion of residents sleeping under LLIN the night before the visit was 2.7%. Indoor residual spray was carried out in 901 (91.7%) households within one year before the visit. Of 349 persons examined 39(11.2%) were positive for malaria parasite. Plasmodium vivax was the dominant (84.6%) cause of malaria infection. The average incidence of malaria episode was 4.6 (varied among villages from 0 to 23.4 episodes) per 10,000 person weeks of observation. Higher malaria incidence was observed among children under 5, 5-14 years and in villages near the lake shore with 6.8, 6.3 and 8 episodes per 10,000 person weeks, respectively. The intra-cluster-correlation coefficient of malaria episodes was 0.27.
Conclusion This pilot study showed that the incidence of malaria was higher in villages near the lake shore and in children less than 15 years old than the average incidence rate. The variation in malaria incidence between villages will be considered to calculate an appropriate sample size for the main trial.